There is an interest in extending the wavelength of active regions on GaAs and GaN. For longwave active regions on GaAs a number of suggestions have been made that fit into three approaches. The first approach relates to introducing nitrogen into an InGaAs quantum well to lower the quantum well bandgap; the second approach relates to using highly strained narrow bandgap GaAsSb quantum wells; and the third approach relates to increasing the strain to the point where quantum dot active regions are formed.
Of the three approaches, the first has been the most successful. However, growth of the nitrogen incorporating quantum wells by metal organic chemical vapor deposition (MOCVD) has been difficult because of the poor nitrogen incorporation into the InGaAs quantum wells. Additionally, the reliability of MOCVD grown InGaAs:N is undetermined because efforts have been primarily directed at molecular beam epitaxy (MBE) based growth techniques where the incorporation of nitrogen into InGaAs is easier.
In the second approach, GaAsSb quantum wells suffer from type II band alignment with GaAs and from the requirement of a low V/III ratio needed to grow antimonide based compounds on GaAs. The requirement of a low V/III ratio typically results in poor quality quantum wells.
In the third approach the low density of the quantum dots results in a low gain and broad spectra due to the randomness of the quantum dot size distribution.
For InGaN active regions on GaN, wavelengths are typically extended from the near ultraviolet to the blue or green portion of the spectrum by increasing the indium content by between about 30 percent to 50 percent. However, the optoelectronic quality of high indium content InGaN is severely degraded resulting in low efficiencies at long wavelengths.